Vibeleaderboard
← Benchmark guide

Model results

LiveBench

Choose a shared evaluation protocol to compare model scores. Versions, sources and configurations stay separate.

Atlas snapshot captured 10/19/2026, 10:20:28 AM UTC. This is an import timestamp, not the evaluation date. Coverage may be incomplete.

Reviewed recommendation and limits

Use LiveBench V2 as a suite-level complement to AA Intelligence Index for a general-purpose shortlist. Start with its Browser Use or Agentic suite for screenshot interaction or persistent app work. Choose Reasoning for ground-truth answer checks.

V1 and V2 2026-10-14 stay separate. V2 equally averages measured suites and uses one run at supported effort. Grading mixes answer checks, tests, application outcomes and rubrics. Imported groups below may use different protocols; this review does not certify each run. Single runs and averages over different measured suites limit overall comparisons. Research uses rubrics and Design uses visual and structural grading; grader validation is not disclosed. V1 contamination claims do not certify V2. Measurement dates are not reported.

Publisher results LiveBench methodology
← All evaluation groups

LiveBench overall

Source
LiveBench
Version
v2
Measurement date
Not reported
Metric
% · Higher is better
Evaluation scope
Release 2026-10-14; observed 2026-10-19; result date not supplied; equal mean of suites: long_agentic, reasoning, coding, browser, data_analysis, research, design; single run; publisher effort settings (highest supported unless noted on the board)

Exact score ties share a position. Positions do not establish statistical significance. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.

  1. #1Claude Fable 5.1

    anthropic/claude-fable-5-1

    87.99%

    View modelEvidence independent

  2. #2Claude Opus 5.5

    anthropic/claude-opus-5-5

    86.53%

    View modelEvidence independent

  3. #3GPT-6 Astra

    openai/gpt-6-astra-20260903

    86.14%

    View modelEvidence independent

  4. #4GPT-6.1 Sol

    openai/gpt-6.1-sol-20260929

    85.41%

    View modelEvidence independent

  5. #5Gemini 4 Argon

    google/gemini-4-argon-20261013

    84.1%

    View modelEvidence independent

  6. #6Claude Fable 5

    anthropic/claude-fable-5

    83.61%

    View modelEvidence independent

  7. #7Claude Sonnet 5.5

    anthropic/claude-sonnet-5-5

    82.46%

    View modelEvidence independent

  8. #8Claude Opus 5

    anthropic/claude-opus-5

    82.09%

    View modelEvidence independent

  9. #9GPT-6 Sol

    openai/gpt-6-sol-20260922

    81.91%

    View modelEvidence independent

  10. #10DeepSeek V4.1 Flash 1017

    deepseek/deepseek-v4.1-flash-20261017

    79.2%

    View modelEvidence independent

  11. #11DeepSeek V4.1 Flash

    deepseek/deepseek-v4.1-flash-20260910

    78.04%

    View modelEvidence independent

  12. #12Claude Opus 4.8

    anthropic/claude-opus-4-8

    76.77%

    View modelEvidence independent

  13. #13GPT-5.6 Sol

    openai/gpt-5.6-sol-20260709

    76.26%

    View modelEvidence independent

  14. #14GPT-5.6 Terra

    openai/gpt-5.6-terra-20260709

    74.99%

    View modelEvidence independent

  15. #15Grok 4.6

    x-ai/grok-4.6-20260810

    74.94%

    View modelEvidence independent

  16. #16MiMo-V2.6-Pro

    xiaomi/mimo-v2.6-pro-20260921

    74.87%

    View modelEvidence independent

  17. #17Muse Spark 1.3

    meta/muse-spark-1.3-20260902

    74.76%

    View modelEvidence independent

  18. #18Claude Sonnet 5

    anthropic/claude-sonnet-5

    74.63%

    View modelEvidence independent

  19. #19Claude Haiku 5.5

    anthropic/claude-haiku-5-5

    74.49%

    View modelEvidence independent

  20. #20Grok 4.7

    x-ai/grok-4.7-20260916

    74.09%

    View modelEvidence independent

  21. #21GLM 5.3 Flash

    z-ai/glm-5.3-flash-20260826

    70.61%

    View modelEvidence independent

  22. #22Grok 4.5

    x-ai/grok-4.5-20260708

    70.06%

    View modelEvidence independent

  23. #23GPT-6 Luna

    openai/gpt-6-luna-20260922

    69.4%

    View modelEvidence independent

  24. #24Qwen3.8 Flash

    qwen/qwen3.8-flash-20260826

    68.79%

    View modelEvidence independent

  25. #25GPT-5.5

    openai/gpt-5.5-20260423

    68.7%

    View modelEvidence independent

  26. #26Gemini 3.8 Flash

    google/gemini-3.8-flash-20260902

    67.34%

    View modelEvidence independent

  27. #27Gemini 3.7 Flash

    google/gemini-3.7-flash-20260813

    67.19%

    View modelEvidence independent

  28. #28GPT-5.6 Luna

    openai/gpt-5.6-luna-20260709

    66.09%

    View modelEvidence independent

  29. #29Claude Sonnet 4.6

    anthropic/claude-sonnet-4-6

    65.09%

    View modelEvidence independent

  30. #30Kimi K3

    moonshotai/kimi-k3-20260715

    63.39%

    View modelEvidence independent

  31. #31Qwen3.8 27B

    qwen/qwen3.8-27b-20260814

    62.66%

    View modelEvidence independent

  32. #32Muse Spark 1.2

    meta/muse-spark-1.2-20260805

    62.4%

    View modelEvidence independent

  33. #33Step 5 Preview

    stepfun/step-5-preview-20261008

    62.01%

    View modelEvidence independent

  34. #34GLM 5.3

    z-ai/glm-5.3-20260816

    60.74%

    View modelEvidence independent

  35. #35Muse Spark Open

    meta/muse-spark-open-20261015

    57.57%

    View modelEvidence independent

  36. #36MiniMax M3

    minimax/minimax-m3-20260531

    53.74%

    View modelEvidence independent

  37. #37Inkling

    thinkingmachines/inkling-20260715

    51.91%

    View modelEvidence independent

  38. #38Mistral Large 4

    mistralai/mistral-large-4-0-20261006

    47.7%

    View modelEvidence independent

  39. #39Claude Haiku 4.5

    anthropic/claude-haiku-4-5

    46.04%

    View modelEvidence independent

  40. #40Gemma 4 31B

    google/gemma-4-31b-it-20260402

    41.53%

    View modelEvidence independent
LiveBench model results | VibeLeaderboard